| openstackgerrit | Fei Long Wang proposed a change to openstack/tempest: Add cases for Swift scenario test https://review.openstack.org/99514 | 00:02 |
|---|---|---|
| openstackgerrit | Fei Long Wang proposed a change to openstack/tempest: Add cases for Swift scenario test https://review.openstack.org/99514 | 00:03 |
| *** aimon has quit IRC | 00:05 | |
| *** mmaglana has quit IRC | 00:05 | |
| *** GMann_ has joined #openstack-qa | 00:06 | |
| *** GMann has quit IRC | 00:06 | |
| *** GMann_ has quit IRC | 00:07 | |
| *** GMann has joined #openstack-qa | 00:07 | |
| openstackrecheck | Console logs not available after 13:43s for check-tempest-dsvm-neutron 94944,12,dc2a78b | 00:08 |
| *** oomichi_sleeping is now known as oomichi | 00:09 | |
| *** dims__ has quit IRC | 00:11 | |
| *** dims has joined #openstack-qa | 00:11 | |
| *** hemna is now known as hemna_ | 00:12 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-dev/devstack: [Fedora] Devstack fails to install in firewalld less system https://review.openstack.org/99047 | 00:16 |
| *** dims has quit IRC | 00:21 | |
| *** dmorita has joined #openstack-qa | 00:21 | |
| *** aimon__ has joined #openstack-qa | 00:22 | |
| openstackrecheck | Console logs not available after 13:38s for check-tempest-dsvm-postgres-full 99460,1,a4c4366 | 00:24 |
| openstackgerrit | Ian Wienand proposed a change to openstack-dev/devstack: [Fedora] Devstack fails to install in firewalld less system https://review.openstack.org/99047 | 00:25 |
| *** matsuhashi has joined #openstack-qa | 00:27 | |
| *** dims has joined #openstack-qa | 00:27 | |
| *** malini1 has joined #openstack-qa | 00:31 | |
| *** yjiang51 has joined #openstack-qa | 00:31 | |
| *** nosnos has joined #openstack-qa | 00:32 | |
| *** yamahata has joined #openstack-qa | 00:33 | |
| openstackrecheck | Console logs not available after 13:26s for gate-nova-python26 99442,4,0f2d516 | 00:40 |
| *** malini1 has quit IRC | 00:44 | |
| *** malini1 has joined #openstack-qa | 00:44 | |
| *** ildikov_ has joined #openstack-qa | 00:48 | |
| *** ildikov has quit IRC | 00:51 | |
| openstackrecheck | Console logs not available after 13:25s for gate-horizon-python26 99464,1,074a01e | 00:54 |
| openstackgerrit | Ghanshyam Mann proposed a change to openstack/tempest: Verify list_instance_action attributes of Nova API https://review.openstack.org/85666 | 00:54 |
| openstackrecheck | Console logs not available after 13:26s for check-tempest-dsvm-neutron 99459,1,d55022a | 01:07 |
| openstackrecheck | Console logs not available after 13:29s for gate-nova-pep8 98529,4,19d75ca | 01:21 |
| *** rbak has quit IRC | 01:27 | |
| *** nati_ueno has quit IRC | 01:30 | |
| openstackgerrit | Ghanshyam Mann proposed a change to openstack/tempest: make 'server' as 'required' in server update schema https://review.openstack.org/99319 | 01:34 |
| openstackrecheck | Console logs not available after 13:27s for check-grenade-dsvm 80847,1,d4f31c5 | 01:34 |
| *** yjiang has joined #openstack-qa | 01:35 | |
| *** wramthun has quit IRC | 01:40 | |
| *** wramthun has joined #openstack-qa | 01:41 | |
| *** rwsu has quit IRC | 01:42 | |
| openstackrecheck | Console logs not available after 13:25s for check-grenade-dsvm 74619,2,1c73371 | 01:47 |
| openstackgerrit | Fei Long Wang proposed a change to openstack/tempest: Add cases for Swift scenario test https://review.openstack.org/99514 | 01:52 |
| *** yjiang51 has quit IRC | 01:54 | |
| *** thomasbiege1 has joined #openstack-qa | 01:55 | |
| *** thomasbiege has quit IRC | 01:58 | |
| openstackrecheck | Console logs not available after 13:31s for check-grenade-dsvm 99106,2,d652799 | 02:01 |
| *** masayukig has quit IRC | 02:10 | |
| *** masayukig has joined #openstack-qa | 02:12 | |
| openstackrecheck | Console logs not available after 13:33s for check-tempest-dsvm-neutron 84244,8,a9f468f | 02:15 |
| *** mmaglana has joined #openstack-qa | 02:28 | |
| openstackrecheck | Console logs not available after 13:44s for check-tempest-dsvm-postgres-full 99351,1,fccfd8d | 02:28 |
| *** malini1 has quit IRC | 02:29 | |
| *** markmcclain has quit IRC | 02:31 | |
| *** mrodden1 has joined #openstack-qa | 02:35 | |
| *** mrodden has quit IRC | 02:35 | |
| *** dims has quit IRC | 02:40 | |
| openstackrecheck | Console logs not available after 13:28s for check-tempest-dsvm-postgres-full 98693,4,76374b7 | 02:42 |
| *** bkopilov_ has quit IRC | 02:48 | |
| openstackrecheck | Console logs not available after 13:25s for check-tempest-dsvm-neutron 69726,12,6209ce2 | 02:55 |
| openstackgerrit | ChangBo Guo(gcb) proposed a change to openstack/tempest: Fix calls to mock.assert_not_called() https://review.openstack.org/93534 | 03:00 |
| *** Madan has joined #openstack-qa | 03:04 | |
| *** coolsvap|afk is now known as coolsvap | 03:04 | |
| *** zhikunliu has joined #openstack-qa | 03:05 | |
| *** dims_ has joined #openstack-qa | 03:07 | |
| openstackrecheck | Console logs not available after 13:24s for check-tempest-dsvm-neutron 99450,1,9c0a53f | 03:09 |
| *** dims_ has quit IRC | 03:12 | |
| *** Longgeek has joined #openstack-qa | 03:21 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-dev/devstack: Use LOCAL_PATH of /tmp for horizon https://review.openstack.org/99559 | 03:21 |
| openstackrecheck | Console logs not available after 13:44s for check-tempest-dsvm-neutron-pg 99437,1,803dfb4 | 03:22 |
| openstackrecheck | Console logs not available after 13:24s for gate-horizon-pep8 99446,2,68f778b | 03:36 |
| *** nosnos has quit IRC | 03:44 | |
| *** yjiang51 has joined #openstack-qa | 03:45 | |
| openstackrecheck | Console logs not available after 13:23s for gate-grenade-dsvm 91634,14,7dfc7fa | 03:49 |
| *** coolsvap is now known as coolsvap|afk | 04:01 | |
| openstackrecheck | Console logs not available after 13:28s for gate-horizon-pep8 96800,2,88632cf | 04:03 |
| *** yfried has quit IRC | 04:08 | |
| *** Poornima has joined #openstack-qa | 04:09 | |
| *** matsuhashi has quit IRC | 04:11 | |
| openstackrecheck | Console logs not available after 13:32s for check-grenade-dsvm 97695,1,2da6d6d | 04:16 |
| *** harlowja is now known as harlowja_away | 04:16 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-dev/devstack: Use LOCAL_PATH of /tmp for horizon https://review.openstack.org/99559 | 04:23 |
| openstackrecheck | Console logs not available after 13:26s for check-tempest-dsvm-neutron 93570,4,04ca188 | 04:30 |
| *** GMann_ has joined #openstack-qa | 04:31 | |
| *** bkopilov_ has joined #openstack-qa | 04:32 | |
| *** GMann has quit IRC | 04:34 | |
| openstackgerrit | Abhijeet Jain proposed a change to openstack/tempest: adds the clients and tests for CINDER QoS V1 API https://review.openstack.org/90306 | 04:41 |
| *** raies has joined #openstack-qa | 04:42 | |
| openstackrecheck | Console logs not available after 13:24s for check-tempest-dsvm-full 97600,5,7ed3b41 | 04:43 |
| *** yfried has joined #openstack-qa | 04:43 | |
| *** Abhijeet_ has joined #openstack-qa | 04:43 | |
| *** yjiang51 has quit IRC | 04:44 | |
| *** matsuhashi has joined #openstack-qa | 04:48 | |
| *** nosnos has joined #openstack-qa | 04:48 | |
| *** Madan has quit IRC | 04:49 | |
| *** markmcclain has joined #openstack-qa | 04:50 | |
| openstackrecheck | Console logs not available after 13:28s for check-tempest-dsvm-full-icehouse 92524,8,9492331 | 04:57 |
| *** psharma has joined #openstack-qa | 04:57 | |
| *** mmaglana has quit IRC | 04:58 | |
| *** mmaglana has joined #openstack-qa | 04:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/tempest: Updated from global requirements https://review.openstack.org/96829 | 05:00 |
| *** yfried_ has joined #openstack-qa | 05:01 | |
| *** yfried has quit IRC | 05:02 | |
| *** coolsvap|afk is now known as coolsvap | 05:02 | |
| *** mmaglana has quit IRC | 05:03 | |
| *** k4n0 has joined #openstack-qa | 05:04 | |
| openstackrecheck | Console logs not available after 13:31s for check-tempest-dsvm-full 99462,1,b22bb19 | 05:10 |
| openstackgerrit | ajay proposed a change to openstack/tempest: Add CLI test of "neutron service-provider-list" https://review.openstack.org/96422 | 05:12 |
| *** Longgeek_ has joined #openstack-qa | 05:21 | |
| *** salv-orlando has joined #openstack-qa | 05:23 | |
| openstackrecheck | Console logs not available after 13:23s for check-grenade-dsvm 95911,2,ae196c1 | 05:24 |
| *** Longgeek has quit IRC | 05:24 | |
| openstackgerrit | A change was merged to openstack/tempest: Add compute notifications tests for ceilometer https://review.openstack.org/64136 | 05:31 |
| *** salv-orlando has quit IRC | 05:33 | |
| openstackrecheck | Console logs not available after 13:27s for check-grenade-dsvm 80358,1,665e378 | 05:37 |
| *** mmaglana has joined #openstack-qa | 05:49 | |
| *** SridharG has joined #openstack-qa | 05:50 | |
| openstackrecheck | Console logs not available after 13:24s for gate-nova-pep8 99473,1,94a2e30 | 05:50 |
| *** GMann_ has quit IRC | 05:54 | |
| *** enikanorov has joined #openstack-qa | 05:58 | |
| *** oomichi has quit IRC | 05:58 | |
| *** yfried_ is now known as yfried | 06:03 | |
| openstackrecheck | Console logs not available after 13:24s for check-tempest-dsvm-postgres-full 97139,7,e718609 | 06:04 |
| openstackgerrit | Ian Wienand proposed a change to openstack-dev/devstack: Use LOCAL_PATH of /tmp for horizon https://review.openstack.org/99559 | 06:08 |
| *** thomasbiege1 has left #openstack-qa | 06:14 | |
| *** markmcclain has quit IRC | 06:15 | |
| openstackrecheck | Console logs not available after 13:27s for check-tempest-dsvm-neutron 98542,3,288df85 | 06:17 |
| *** markmcclain has joined #openstack-qa | 06:18 | |
| *** ildikov_ has quit IRC | 06:20 | |
| openstackrecheck | Console logs not available after 13:25s for gate-heat-pep8 99479,1,ae8879e | 06:31 |
| *** Mike77 has joined #openstack-qa | 06:36 | |
| *** ildikov has joined #openstack-qa | 06:41 | |
| openstackrecheck | Console logs not available after 13:28s for gate-heat-pep8 99470,2,82aff5e | 06:44 |
| *** jlibosva has joined #openstack-qa | 06:54 | |
| *** GMann has joined #openstack-qa | 06:57 | |
| *** [1]evgenyf has joined #openstack-qa | 06:57 | |
| openstackrecheck | Console logs not available after 13:27s for check-tempest-dsvm-neutron 97884,8,85eb031 | 06:58 |
| *** GMann has quit IRC | 07:02 | |
| openstackgerrit | Marc Koderer proposed a change to openstack/qa-specs: Add spec for bp:/fuzzy-test https://review.openstack.org/99585 | 07:04 |
| *** oomichi has joined #openstack-qa | 07:04 | |
| openstackgerrit | Marc Koderer proposed a change to openstack/qa-specs: Add spec for bp:/fuzzy-test https://review.openstack.org/99585 | 07:05 |
| *** ildikov has quit IRC | 07:07 | |
| *** morrison has joined #openstack-qa | 07:07 | |
| *** safchain has joined #openstack-qa | 07:08 | |
| *** morrison has quit IRC | 07:09 | |
| *** markmcclain has quit IRC | 07:11 | |
| openstackrecheck | Console logs not available after 13:32s for check-grenade-dsvm 98624,3,7be8d8c | 07:11 |
| *** yfried has quit IRC | 07:12 | |
| *** yfried has joined #openstack-qa | 07:12 | |
| *** yfried_ has joined #openstack-qa | 07:18 | |
| *** yfried has quit IRC | 07:19 | |
| *** yfried_ is now known as yfried | 07:19 | |
| *** zhikunliu1 has joined #openstack-qa | 07:21 | |
| *** zhikunliu has quit IRC | 07:23 | |
| *** ildikov has joined #openstack-qa | 07:24 | |
| openstackrecheck | Console logs not available after 13:23s for check-tempest-dsvm-full-icehouse 99231,2,119d989 | 07:25 |
| *** zhikunliu1 has quit IRC | 07:26 | |
| *** zhikunliu has joined #openstack-qa | 07:31 | |
| *** yrabl has joined #openstack-qa | 07:34 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack-dev/devstack: Don't exit if we can't restart firewalld https://review.openstack.org/99590 | 07:35 |
| *** mpavlase has joined #openstack-qa | 07:37 | |
| yjiang | hi folks,I use "Gerrit trigger" to trigger Openstack 3rd part test.I've set Verify values "Successful 1", "Failed -1".My CI test could be triggered as normal,but the "Verified +/-1" will not be shown on the review page after CI test is done.Why this happened?Is there anyone could help to answer my question?Any suggestion is appreciated.Thx a lot! | 07:37 |
| openstackrecheck | Console logs not available after 13:25s for check-tempest-dsvm-neutron 86131,12,45788e8 | 07:38 |
| chmouel | whoever takes care of the Broacade CI may be interested to know that it seems to be broken atm: https://review.openstack.org/#/c/99590/ (cannot merge/rebase patches) | 07:43 |
| *** leseb has joined #openstack-qa | 07:45 | |
| *** mmaglana has quit IRC | 07:48 | |
| *** StevenK has quit IRC | 07:50 | |
| *** qba73 has joined #openstack-qa | 07:51 | |
| *** StevenK has joined #openstack-qa | 07:51 | |
| openstackrecheck | Console logs not available after 13:25s for check-tempest-dsvm-neutron 96094,5,822b4bc | 07:51 |
| *** leseb has quit IRC | 07:56 | |
| *** leseb has joined #openstack-qa | 07:57 | |
| *** leseb has quit IRC | 08:01 | |
| *** jordanP has joined #openstack-qa | 08:01 | |
| *** oomichi has quit IRC | 08:05 | |
| *** Mike77 has quit IRC | 08:09 | |
| openstackrecheck | Console logs not available after 13:36s for check-tempest-dsvm-neutron 96345,3,a1721b0 | 08:10 |
| *** leseb has joined #openstack-qa | 08:12 | |
| *** enikanorov__ has quit IRC | 08:12 | |
| *** fbo_away is now known as fbo | 08:14 | |
| *** lucasagomes has joined #openstack-qa | 08:14 | |
| *** plomakin_ has quit IRC | 08:16 | |
| *** asvechnikov has quit IRC | 08:16 | |
| *** vrovachev has quit IRC | 08:16 | |
| *** Poornima has quit IRC | 08:17 | |
| *** smurashov has quit IRC | 08:17 | |
| *** eezhova has quit IRC | 08:17 | |
| *** ilyashakhat has quit IRC | 08:17 | |
| *** gszasz has joined #openstack-qa | 08:18 | |
| *** gszasz has quit IRC | 08:18 | |
| *** gszasz has joined #openstack-qa | 08:18 | |
| *** locke105 has quit IRC | 08:19 | |
| *** gszasz has quit IRC | 08:19 | |
| *** gszasz has joined #openstack-qa | 08:20 | |
| openstackrecheck | Console logs not available after 13:34s for check-grenade-dsvm-partial-ncpu 98899,4,46ae754 | 08:23 |
| *** giulivo has joined #openstack-qa | 08:28 | |
| openstackrecheck | Console logs not available after 13:28s for gate-trove-python26 94484,7,50d7649 | 08:37 |
| *** zhikunliu has quit IRC | 08:38 | |
| *** afazekas has joined #openstack-qa | 08:42 | |
| openstackgerrit | James Chapman proposed a change to openstack-dev/devstack: This feature allows the deployer define the type of OVS datapath. https://review.openstack.org/99447 | 08:42 |
| *** dims_ has joined #openstack-qa | 08:49 | |
| openstackrecheck | Console logs not available after 14:16s for check-grenade-dsvm 97082,2,8dcde22 | 08:52 |
| openstackrecheck | openstack/ceilometer change: https://review.openstack.org/97702 failed because of: gate-tempest-dsvm-neutron: https://bugs.launchpad.net/bugs/1306029 https://bugs.launchpad.net/bugs/1328983 | 08:52 |
| openstackgerrit | James Chapman proposed a change to openstack-dev/devstack: Add OVS_DATAPATH_TYPE to select OVS datapath type. https://review.openstack.org/99447 | 08:53 |
| *** tkammer has joined #openstack-qa | 08:53 | |
| *** dims_ has quit IRC | 08:54 | |
| *** coolsvap is now known as coolsvap|afk | 08:55 | |
| *** zhikunliu has joined #openstack-qa | 09:01 | |
| *** zhikunliu has quit IRC | 09:01 | |
| *** zhikunliu has joined #openstack-qa | 09:02 | |
| *** luqas has quit IRC | 09:06 | |
| *** coolsvap|afk is now known as coolsvap | 09:06 | |
| openstackrecheck | Console logs not available after 13:32s for check-tempest-dsvm-neutron 96618,1,ad5c083 | 09:08 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Set default value for IPv6 subnet attribute flag https://review.openstack.org/97383 | 09:22 |
| openstackrecheck | Console logs not available after 15:26s for check-tempest-dsvm-neutron 97946,2,b6d458f | 09:25 |
| *** tosky has joined #openstack-qa | 09:36 | |
| *** GMann has joined #openstack-qa | 09:37 | |
| *** luqas has joined #openstack-qa | 09:38 | |
| openstackrecheck | Console logs not available after 13:27s for gate-heat-pep8 98533,4,edcc31b | 09:42 |
| *** dmorita has quit IRC | 09:46 | |
| openstackgerrit | afazekas proposed a change to openstack-dev/devstack: selinux permissive on Fedora https://review.openstack.org/99559 | 09:53 |
| *** oomichi has joined #openstack-qa | 09:54 | |
| *** Manishanker has joined #openstack-qa | 09:54 | |
| openstackrecheck | Console logs not available after 13:32s for check-tempest-dsvm-postgres-full 93570,4,21afefd | 09:56 |
| *** zhikunliu has quit IRC | 10:05 | |
| openstackgerrit | Ghanshyam Mann proposed a change to openstack/tempest: Cleanup for local variable in test_attach_volume https://review.openstack.org/97385 | 10:07 |
| openstackrecheck | Console logs not available after 13:23s for check-tempest-dsvm-postgres-full 89427,2,61f08c8 | 10:11 |
| *** asvechnikov has joined #openstack-qa | 10:11 | |
| *** smurashov has joined #openstack-qa | 10:12 | |
| *** leseb has quit IRC | 10:14 | |
| *** leseb has joined #openstack-qa | 10:15 | |
| *** leseb has quit IRC | 10:19 | |
| *** matsuhashi has quit IRC | 10:20 | |
| *** matsuhashi has joined #openstack-qa | 10:20 | |
| *** yamahata has quit IRC | 10:21 | |
| openstackgerrit | Andrea Frittoli proposed a change to openstack/qa-specs: add bp:test-accounts https://review.openstack.org/98400 | 10:21 |
| *** dims_ has joined #openstack-qa | 10:23 | |
| *** matsuhashi has quit IRC | 10:25 | |
| openstackrecheck | Console logs not available after 15:09s for check-tempest-dsvm-neutron 96399,1,37f02cf | 10:26 |
| *** Abhijeet_ has quit IRC | 10:30 | |
| openstackgerrit | Andrea Frittoli proposed a change to openstack/qa-specs: add bp:test-accounts https://review.openstack.org/98400 | 10:33 |
| openstackgerrit | A change was merged to openstack/tempest: Sahara: add API tests for cluster templates https://review.openstack.org/90101 | 10:38 |
| openstackgerrit | A change was merged to openstack/tempest: adds the missing _interface variable https://review.openstack.org/97756 | 10:38 |
| *** kmartin has quit IRC | 10:38 | |
| openstackgerrit | A change was merged to openstack/tempest: Add veritification to name property in cinder volume https://review.openstack.org/95714 | 10:38 |
| openstackrecheck | Console logs not available after 13:31s for check-tempest-dsvm-neutron 96280,3,a7f2c7a | 10:39 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/98864 failed because of: gate-tempest-dsvm-postgres-full: https://bugs.launchpad.net/bugs/1254872 | 10:40 |
| openstackgerrit | Andrea Frittoli proposed a change to openstack/qa-specs: add bp:test-accounts https://review.openstack.org/98400 | 10:40 |
| *** thomasbiege has joined #openstack-qa | 10:42 | |
| andreaf | dkranz, sdague, mtreinish: ^^^ the spec is not complete yet - I'm missing the bit about how to sync account locking across different tempest processes - but I would appreciate an early feedback if possible | 10:43 |
| *** lucasagomes has left #openstack-qa | 10:43 | |
| *** Madan has joined #openstack-qa | 10:43 | |
| *** leseb has joined #openstack-qa | 10:46 | |
| *** leseb has quit IRC | 10:51 | |
| *** mkollaro has joined #openstack-qa | 10:52 | |
| *** Poornima has joined #openstack-qa | 10:53 | |
| openstackrecheck | Console logs not available after 13:33s for check-tempest-dsvm-full 94668,5,33d861c | 10:55 |
| *** dkorn has joined #openstack-qa | 10:58 | |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/devstack: convert devstack from awk to outfilter https://review.openstack.org/99399 | 10:59 |
| *** Madan has quit IRC | 11:00 | |
| *** atiwari has quit IRC | 11:04 | |
| *** gszasz has quit IRC | 11:08 | |
| openstackrecheck | Console logs not available after 13:23s for gate-cinder-pep8 99501,2,446ec08 | 11:08 |
| *** nosnos has quit IRC | 11:13 | |
| openstackrecheck | Console logs not available after 14:53s for check-tempest-dsvm-neutron 99470,3,598ea2b | 11:23 |
| *** evgenyf has joined #openstack-qa | 11:26 | |
| *** yjiang has quit IRC | 11:28 | |
| *** [1]evgenyf has quit IRC | 11:28 | |
| *** plomakin has joined #openstack-qa | 11:29 | |
| *** leseb has joined #openstack-qa | 11:30 | |
| *** k4n0 has quit IRC | 11:31 | |
| *** jordanP has quit IRC | 11:32 | |
| *** luqas has quit IRC | 11:36 | |
| *** coolsvap is now known as coolsvap|afk | 11:36 | |
| openstackrecheck | Console logs not available after 13:22s for check-tempest-dsvm-neutron 98624,3,57e4567 | 11:37 |
| *** jordanP has joined #openstack-qa | 11:37 | |
| *** GMann has quit IRC | 11:44 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: gate-tempest-dsvm-virtual-ironic is in our gate https://review.openstack.org/99635 | 11:46 |
| openstackrecheck | Console logs not available after 13:26s for check-tempest-dsvm-full 97385,1,28dc1a9 | 11:51 |
| *** yfried_ has joined #openstack-qa | 11:53 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: gate-tempest-dsvm-virtual-ironic is in our gate https://review.openstack.org/99635 | 11:54 |
| *** thomasbiege has quit IRC | 11:55 | |
| *** yfried has quit IRC | 11:56 | |
| *** lbragstad has quit IRC | 11:59 | |
| *** thomasbiege has joined #openstack-qa | 12:00 | |
| openstackrecheck | Console logs not available after 13:28s for check-tempest-dsvm-neutron 48850,13,57eb411 | 12:05 |
| *** openstackrecheck has quit IRC | 12:08 | |
| *** openstackrecheck has joined #openstack-qa | 12:08 | |
| *** yamahata has joined #openstack-qa | 12:10 | |
| EmilienM | sdague: I think you should use your power to approve https://review.openstack.org/#/c/99412/ - it will block ceilometer patchs in the gate | 12:13 |
| EmilienM | it will avoid to block* | 12:13 |
| sdague | EmilienM: done | 12:16 |
| *** dkranz has joined #openstack-qa | 12:19 | |
| *** adalbas has joined #openstack-qa | 12:20 | |
| EmilienM | sdague: thx | 12:21 |
| *** dims_ has quit IRC | 12:22 | |
| *** dims_ has joined #openstack-qa | 12:23 | |
| *** aysyd has joined #openstack-qa | 12:23 | |
| openstackgerrit | David Shrewsbury proposed a change to openstack/tempest: Add ironic instance rebuild advanced test https://review.openstack.org/94439 | 12:24 |
| *** weshay has joined #openstack-qa | 12:25 | |
| *** dims_ is now known as dims | 12:25 | |
| openstackrecheck | Console logs not available after 16:00s for gate-nova-pep8 99623,1,1fd63e5 | 12:28 |
| *** amotoki has quit IRC | 12:31 | |
| openstackgerrit | Andrey Pavlov proposed a change to openstack/tempest: fix usage of allow_tenant_isolation https://review.openstack.org/97139 | 12:34 |
| *** Manishanker has quit IRC | 12:38 | |
| *** GMann has joined #openstack-qa | 12:40 | |
| *** psharma has quit IRC | 12:41 | |
| *** chandan_kumar has joined #openstack-qa | 12:41 | |
| openstackrecheck | Console logs not available after 13:27s for check-tempest-dsvm-postgres-full 99624,1,bf565d4 | 12:42 |
| *** thomasbiege has left #openstack-qa | 12:42 | |
| *** gszasz has joined #openstack-qa | 12:44 | |
| *** chandan_kumar has quit IRC | 12:44 | |
| *** coolsvap|afk has quit IRC | 12:46 | |
| *** openstackgerrit has quit IRC | 12:46 | |
| *** yfried_ has quit IRC | 12:46 | |
| *** coolsvap|afk has joined #openstack-qa | 12:46 | |
| *** ekhugen has joined #openstack-qa | 12:46 | |
| *** GMann has quit IRC | 12:47 | |
| andreaf | dkranz: ping | 12:47 |
| dkranz | andreaf: hi | 12:47 |
| andreaf | dkranz: hi | 12:48 |
| *** openstackgerrit has joined #openstack-qa | 12:48 | |
| andreaf | dkranz: I started the spec for test accounts: https://review.openstack.org/98400 - it's not complete yet, but I'd like an early feedback if possible | 12:48 |
| dkranz | andreaf: sure | 12:48 |
| andreaf | dkranz: thanks | 12:49 |
| *** oomichi has quit IRC | 12:49 | |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/tempest: Expand baremetal nodes coverage `POST` https://review.openstack.org/74065 | 12:49 |
| *** chandankumar has joined #openstack-qa | 12:52 | |
| EmilienM | sdague: it's strange to see the check-grenade-dsvm working without the dbsync stuff, am I missing something ? > https://review.openstack.org/#/c/99634/ | 12:52 |
| openstackrecheck | Console logs not available after 13:27s for gate-tempest-dsvm-full 99396,1,9d3af7b | 12:55 |
| *** dims_ has joined #openstack-qa | 12:55 | |
| *** tkammer has quit IRC | 12:56 | |
| *** dims has quit IRC | 12:56 | |
| *** bkopilov_ has quit IRC | 12:56 | |
| *** Poornima has quit IRC | 12:56 | |
| *** lbragstad has joined #openstack-qa | 12:57 | |
| *** luqas has joined #openstack-qa | 13:01 | |
| *** yamahata has quit IRC | 13:03 | |
| *** yamahata has joined #openstack-qa | 13:03 | |
| *** qba73 has quit IRC | 13:06 | |
| openstackrecheck | Console logs not available after 13:23s for check-grenade-dsvm 86257,4,4f02a27 | 13:09 |
| *** mriedem has joined #openstack-qa | 13:12 | |
| *** dkranz has quit IRC | 13:14 | |
| *** tkammer has joined #openstack-qa | 13:17 | |
| *** tkammer has quit IRC | 13:19 | |
| *** mkollaro has quit IRC | 13:22 | |
| *** oomichi has joined #openstack-qa | 13:22 | |
| openstackrecheck | Console logs not available after 13:29s for gate-ceilometer-python26 98834,3,37ea9a1 | 13:22 |
| *** tkammer has joined #openstack-qa | 13:27 | |
| openstackrecheck | Console logs not available after 13:43s for check-grenade-dsvm 99399,2,2d09a07 | 13:36 |
| *** mwagner_lap has quit IRC | 13:36 | |
| *** mkollaro has joined #openstack-qa | 13:40 | |
| *** mlavalle has joined #openstack-qa | 13:42 | |
| *** mlavalle has quit IRC | 13:42 | |
| *** malini1 has joined #openstack-qa | 13:42 | |
| *** sileht has quit IRC | 13:43 | |
| *** mlavalle has joined #openstack-qa | 13:43 | |
| *** zhikunliu has joined #openstack-qa | 13:44 | |
| *** Obulpathi has joined #openstack-qa | 13:48 | |
| *** Obulpathi has joined #openstack-qa | 13:48 | |
| openstackrecheck | Console logs not available after 13:25s for check-tempest-dsvm-full 97831,10,b8f87ab | 13:49 |
| *** malini1 has left #openstack-qa | 13:51 | |
| *** beekneemech is now known as bnemec | 13:54 | |
| *** yfried_ has joined #openstack-qa | 13:55 | |
| *** Longgeek has joined #openstack-qa | 13:55 | |
| *** yfried_ has quit IRC | 13:56 | |
| *** reaper has joined #openstack-qa | 13:56 | |
| *** yfried_ has joined #openstack-qa | 13:56 | |
| *** Longgeek_ has quit IRC | 13:57 | |
| *** malini1 has joined #openstack-qa | 13:59 | |
| *** sileht has joined #openstack-qa | 14:02 | |
| *** yjiang51 has joined #openstack-qa | 14:02 | |
| openstackrecheck | Console logs not available after 13:23s for gate-ceilometer-python26 99140,1,c56a7c7 | 14:03 |
| *** samuelbercovici has joined #openstack-qa | 14:05 | |
| openstackgerrit | Andrea Frittoli proposed a change to openstack/qa-specs: add bp:test-accounts https://review.openstack.org/98400 | 14:06 |
| andreaf | sdague, mtreinish : proposal on how to implement a preconfigured list of accounts in tempest ^^^ | 14:08 |
| *** rbak has joined #openstack-qa | 14:13 | |
| mkoderer | andreaf: nice :) | 14:14 |
| openstackrecheck | Console logs not available after 13:23s for gate-python-swiftclient-python33 99632,1,ec244f1 | 14:16 |
| *** atiwari has joined #openstack-qa | 14:18 | |
| *** davlaps has joined #openstack-qa | 14:24 | |
| *** malini1 has quit IRC | 14:25 | |
| *** malini1 has joined #openstack-qa | 14:25 | |
| *** malini1 has quit IRC | 14:26 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for ceilometer test_notify_alarm bug 1321826 https://review.openstack.org/98149 | 14:27 |
| *** malini1 has joined #openstack-qa | 14:27 | |
| openstackrecheck | Console logs not available after 13:25s for check-tempest-dsvm-full 99370,4,05e9f24 | 14:29 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack-dev/devstack: Ironic: Setup/Configure iPXE https://review.openstack.org/99677 | 14:31 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack-dev/devstack: Ironic: Setup/Configure iPXE https://review.openstack.org/99677 | 14:32 |
| *** coolsvap|afk is now known as coolsvap | 14:33 | |
| *** bkopilov_ has joined #openstack-qa | 14:34 | |
| *** openstackrecheck has quit IRC | 14:37 | |
| *** openstackrecheck has joined #openstack-qa | 14:37 | |
| *** kgriffs|afk is now known as kgriffs | 14:40 | |
| *** yamahata has quit IRC | 14:41 | |
| *** leseb has quit IRC | 14:42 | |
| *** leseb has joined #openstack-qa | 14:42 | |
| *** leseb has quit IRC | 14:47 | |
| *** coolsvap is now known as coolsvap|afk | 14:49 | |
| *** kmartin has joined #openstack-qa | 14:55 | |
| *** oomichi has quit IRC | 14:58 | |
| *** marun has joined #openstack-qa | 14:58 | |
| *** yjiang51 has quit IRC | 14:59 | |
| *** leseb has joined #openstack-qa | 14:59 | |
| *** jlibosva has quit IRC | 15:01 | |
| *** samuelbercovici1 has joined #openstack-qa | 15:01 | |
| *** [1]evgenyf has joined #openstack-qa | 15:02 | |
| *** jlibosva has joined #openstack-qa | 15:02 | |
| *** [3]evgenyf has joined #openstack-qa | 15:04 | |
| *** tkammer has quit IRC | 15:04 | |
| *** evgenyf has quit IRC | 15:04 | |
| *** [3]evgenyf is now known as evgenyf | 15:04 | |
| *** samuelbercovici has quit IRC | 15:04 | |
| *** yrabl has quit IRC | 15:05 | |
| *** [1]evgenyf has quit IRC | 15:07 | |
| *** samuelbercovici1 has quit IRC | 15:07 | |
| *** samuelbercovici has joined #openstack-qa | 15:08 | |
| *** coolsvap|afk is now known as coolsvap | 15:08 | |
| *** andreaf has quit IRC | 15:16 | |
| *** zhikunliu has quit IRC | 15:19 | |
| openstackgerrit | David Shrewsbury proposed a change to openstack/tempest: Add ironic instance rebuild advanced test https://review.openstack.org/94439 | 15:19 |
| *** markmcclain has joined #openstack-qa | 15:22 | |
| *** dkranz has joined #openstack-qa | 15:23 | |
| *** samuelbercovici1 has joined #openstack-qa | 15:28 | |
| *** [1]evgenyf has joined #openstack-qa | 15:28 | |
| *** kgriffs is now known as kgriffs|afk | 15:30 | |
| *** evgenyf has quit IRC | 15:30 | |
| *** [1]evgenyf is now known as evgenyf | 15:30 | |
| *** samuelbercovici has quit IRC | 15:31 | |
| *** samuelbercovici1 is now known as samuelbercovici | 15:31 | |
| *** kgriffs|afk is now known as kgriffs | 15:33 | |
| *** samuelbercovici1 has joined #openstack-qa | 15:33 | |
| *** [1]evgenyf has joined #openstack-qa | 15:33 | |
| openstackgerrit | afazekas proposed a change to openstack/tempest: Heat Overlapping ip issue https://review.openstack.org/99694 | 15:34 |
| *** evgenyf has quit IRC | 15:35 | |
| *** samuelbercovici has quit IRC | 15:36 | |
| morganfainberg | sdague, i'm happy to change what we call the token obscuring, but I am concerned that we can't say it's SHA1 (exclusively) maybe SSHA (salted)? only because if keystone is configured to hash PKI tokens as SHA1, you would actually be able to use hat SHA1 token to "do things" | 15:37 |
| sdague | morganfainberg: ok, color me confused | 15:37 |
| afazekas | sdague, mtreinish, dkranz, stevebaker. mkoderer: can you do a quick review on https://review.openstack.org/99694 | 15:37 |
| morganfainberg | sdague, keystone hashes PKI tokens to the "short-id" (looks like UUID, MD5 was the default). This short ID was used for token revocation list and if you wanted the PKI token to be interchangable with a UUID token | 15:38 |
| *** [1]evgenyf has quit IRC | 15:38 | |
| morganfainberg | sdague, so this "short-id" can be used just like a UUID token. | 15:38 |
| morganfainberg | sdague, recently (a couple weeks ago) we landed a patch to make this hashing not MD5, but any hashing hashlib can handle | 15:39 |
| *** samuelbercovici1 has quit IRC | 15:39 | |
| morganfainberg | if you (say) told keystone to hash PKI tokens to the short-id using SHA1, you now could use the "hashed" debug output just like a normal token. | 15:39 |
| sdague | afazekas: you have heat folks on the review? unfortunately heat templates are pretty domain expert area | 15:39 |
| sdague | afazekas: also we should get rid of hardcoded google dns server in template as well | 15:40 |
| afazekas | sdague: we can keep the hardcoded dns | 15:40 |
| sdague | afazekas: does it ever try to use dns? | 15:40 |
| afazekas | no | 15:40 |
| sdague | because if it does, we really can't | 15:40 |
| sdague | afazekas: it starts a guest | 15:41 |
| sdague | the guest might try to use dns | 15:41 |
| sdague | morganfainberg: so what you are saying is that the SHA1 value would actually be valid? | 15:41 |
| afazekas | sdague: AFAIK in the devstack neutron setups the VM's cannot reach outside network | 15:41 |
| morganfainberg | sdague, under certain configurations of keystone. | 15:42 |
| afazekas | We do not have the iptables masquerade trick | 15:42 |
| sdague | afazekas: if we are fixing it, we should fix that part as well | 15:42 |
| sdague | meaning the hardcoded dns | 15:42 |
| sdague | because otherwise we'll be fixing it later | 15:42 |
| sdague | morganfainberg: does that not seem problematic to you? | 15:42 |
| afazekas | sdague: 8.8.8.8 address belongs to google, no-one expected to use 8.8.8.8 address on his host | 15:43 |
| sdague | afazekas: please also fix that part | 15:43 |
| morganfainberg | sdague, if we salt the hashing, say take the last X bytes of the token and use that as well to hash (at the front?) it would still generate a consistent hash, just not a usable one | 15:43 |
| morganfainberg | sdague, we just can't do "plain" sha1 hasing in debug log. | 15:43 |
| sdague | morganfainberg: it seems like the short id which can actually be used on the wire should be the salted one, no? | 15:44 |
| sdague | I guess that's the part that confuses me | 15:44 |
| sdague | straight hashing is fine for integrity | 15:45 |
| sdague | which is what the debug output is about | 15:45 |
| sdague | however it's not ok for security | 15:45 |
| afazekas | sdague: What do you expect as a fix ? Not configuring dns, or making it as config option ? (at the moment we are using IPs so, dns never not involved) | 15:45 |
| sdague | afazekas: making it a template var | 15:45 |
| sdague | like you did with ip | 15:45 |
| sdague | so it will get passed by the test case | 15:45 |
| afazekas | sdague: ok | 15:46 |
| *** samuelbercovici has joined #openstack-qa | 15:46 | |
| morganfainberg | sdague, the whole concepts revolves around PKI tokens being a bearer token. if you have that, you already have the credentials needed. - we just provided a way to use a "short id" if you wanted to hash the pki token (e.g. make it look and act like a UUID token) - interop story | 15:46 |
| morganfainberg | sdague, i kindof wish we hadn't done that. | 15:47 |
| morganfainberg | sdague, but the long and the short is a bearer token is still a bearer token. hashing+salt doesn't make it less of a bearer token (you could use the full PKI or the short, same effect) | 15:47 |
| sdague | ok, so it just seems weird that we have to salt the debug because unsalted could give you resource access :) | 15:49 |
| morganfainberg | sdague, the short id is also the PK in the SQL token table (or in memcached) | 15:49 |
| sdague | also, this is going to now have to be documented extensively so people can actually debug it, as it's no longer a well understood hash | 15:49 |
| morganfainberg | if that helps with the implementation details. | 15:49 |
| morganfainberg | yep :( | 15:50 |
| sdague | yeh, it just seems terrible to use straight hashing for that | 15:50 |
| morganfainberg | it's annoying is what it is. | 15:50 |
| morganfainberg | well if we use some hashing that isn't hashlib supported :P | 15:50 |
| sdague | and seems like there is probably an exposure in bearer token down the road | 15:50 |
| morganfainberg | the hashing of the PKI token is no different than the UUID tokens were. | 15:51 |
| morganfainberg | so i see why it was supported when we did the move to PKI | 15:51 |
| sdague | it seems like the bearer tokens should be a properly secure hashed, multi round | 15:52 |
| morganfainberg | sdague, what benefit does that provide? | 15:52 |
| morganfainberg | the short-id version keystone needs to be able to lookup what comes inot the API | 15:52 |
| morganfainberg | so it needs to be a key we can act on, the data no longer exists in the token | 15:53 |
| sdague | you can recompute it | 15:53 |
| morganfainberg | PKI tokens provide all data so keystone doesn't need to be asked "hey what is in this token" | 15:53 |
| morganfainberg | sdague, recompute it from what? | 15:53 |
| sdague | so I apparently don't understand the flow :) | 15:54 |
| morganfainberg | sdague, ah, ok | 15:54 |
| sdague | anyway, it seems like the fact that we can't use simple hashing in debug output because that will also give you access to systems | 15:54 |
| sdague | is a fundamental flaw | 15:54 |
| sdague | that might not be exploitable today | 15:54 |
| sdague | but will be | 15:54 |
| sdague | and you'll be redoing it down the road | 15:55 |
| morganfainberg | sdague, the basic idea is if you already have the PKI bearer token, you could create the hash. | 15:55 |
| sdague | ok, you'll have to draw pictures. | 15:55 |
| morganfainberg | sec. | 15:56 |
| morganfainberg | have a quick explination for you | 15:56 |
| sdague | If you think the right thing to do is update all the clients, and document the hell out of this, instead of changing the keystone side to not use a single hash with no seed, so be it | 15:56 |
| morganfainberg | sdague, if we didn't support UUID tokens it would be easy. | 15:56 |
| *** samuelbercovici1 has joined #openstack-qa | 15:57 | |
| *** markmcclain has quit IRC | 15:57 | |
| sdague | I'll let you drive the rest of it on the list | 15:57 |
| sdague | because I definitely don't have time to take this thread any further | 15:57 |
| *** markmcclain has joined #openstack-qa | 15:57 | |
| sdague | and it still feels like an eventual keystone sec hole | 15:57 |
| *** markmcclain has quit IRC | 15:57 | |
| sdague | but that's only my medium level of crypto understanding speaking | 15:58 |
| *** markmcclain has joined #openstack-qa | 15:58 | |
| sdague | but please put a full explanation back on the list, because we should have very clear reasons documented on this | 15:58 |
| morganfainberg | sdague, ah, don't think of bearer tokens as "crypto" they really aren't. short id tokens ask keystone for validation (Every time), pki tokens can be verified w/o asking keystone because they are signed. but i'm gonna go be grumpy at some security folks for this i think. | 15:59 |
| morganfainberg | and then i'll deal with continuing the thread on the ML | 15:59 |
| morganfainberg | so we can try and resolve this some sane way vs. needing silly extra hard work | 15:59 |
| sdague | yeh, given that we're talking about 3rd redo on a bunch of these patches already | 16:00 |
| *** samuelbercovici has quit IRC | 16:00 | |
| sdague | yeh, it just seems like the bearer tokens should still use a more secure hashing technique which isn't guessable | 16:01 |
| sdague | for replay attack purposes | 16:01 |
| sdague | but then again, I'm not an expert there | 16:02 |
| sdague | so be it :) | 16:02 |
| *** samuelbercovici1 has quit IRC | 16:02 | |
| *** mestery has quit IRC | 16:03 | |
| *** mestery has joined #openstack-qa | 16:04 | |
| openstackgerrit | afazekas proposed a change to openstack/tempest: heat client: stop logging to tty https://review.openstack.org/99701 | 16:05 |
| *** samuelbercovici has joined #openstack-qa | 16:06 | |
| *** mwagner_lap has joined #openstack-qa | 16:07 | |
| *** yfried_ has quit IRC | 16:07 | |
| *** aimon__ has quit IRC | 16:09 | |
| *** torandu has quit IRC | 16:11 | |
| *** yfried_ has joined #openstack-qa | 16:11 | |
| *** dustins has joined #openstack-qa | 16:11 | |
| *** torandu has joined #openstack-qa | 16:12 | |
| *** locke105 has joined #openstack-qa | 16:12 | |
| *** mrodden1 has quit IRC | 16:12 | |
| *** leseb has quit IRC | 16:17 | |
| *** markmcclain has quit IRC | 16:23 | |
| *** davlaps is now known as devlaps | 16:25 | |
| *** jordanP has quit IRC | 16:26 | |
| mriedem | i don't think i've spent so much time in the mailing list as this morning in probably forever... | 16:29 |
| mriedem | so many fun topics | 16:29 |
| *** Longgeek has quit IRC | 16:29 | |
| openstackgerrit | afazekas proposed a change to openstack/tempest: Heat Overlapping ip issue https://review.openstack.org/99694 | 16:29 |
| afazekas | sdague: ^^ BTW the overlapping Is the leading cause of the ``waitcondition`` bug at the moment | 16:30 |
| sdague | afazekas: ok, good to know. | 16:30 |
| *** mrodden has joined #openstack-qa | 16:31 | |
| *** safchain has quit IRC | 16:32 | |
| *** openstackgerrit has quit IRC | 16:38 | |
| *** samuelbercovici has quit IRC | 16:44 | |
| *** markmcclain has joined #openstack-qa | 16:44 | |
| *** leseb has joined #openstack-qa | 16:48 | |
| *** leseb has quit IRC | 16:53 | |
| *** harlowja_away is now known as harlowja | 16:54 | |
| *** luqas has quit IRC | 16:54 | |
| *** aimon has joined #openstack-qa | 16:59 | |
| *** rwsu has joined #openstack-qa | 17:00 | |
| *** dustins has quit IRC | 17:01 | |
| *** jlibosva has quit IRC | 17:01 | |
| *** jlibosva has joined #openstack-qa | 17:02 | |
| *** dustins has joined #openstack-qa | 17:04 | |
| *** nati_ueno has joined #openstack-qa | 17:06 | |
| *** tosky has quit IRC | 17:14 | |
| *** jlibosva has quit IRC | 17:23 | |
| *** markmcclain has quit IRC | 17:25 | |
| *** rwsu has quit IRC | 17:30 | |
| *** rwsu has joined #openstack-qa | 17:32 | |
| *** markmcclain has joined #openstack-qa | 17:33 | |
| *** markmcclain has quit IRC | 17:33 | |
| *** markmcclain has joined #openstack-qa | 17:35 | |
| *** lbragstad has quit IRC | 17:39 | |
| devananda | your bot seems to be dead | 17:45 |
| *** afazekas has quit IRC | 17:47 | |
| *** mkollaro has quit IRC | 17:47 | |
| *** mpavlase has quit IRC | 17:48 | |
| *** catohornet has joined #openstack-qa | 17:48 | |
| *** dkorn has quit IRC | 17:48 | |
| *** catohornet has quit IRC | 17:48 | |
| *** lbragstad has joined #openstack-qa | 17:49 | |
| *** leseb has joined #openstack-qa | 17:49 | |
| *** mkollaro has joined #openstack-qa | 17:54 | |
| *** leseb has quit IRC | 17:54 | |
| *** markmcclain has quit IRC | 18:01 | |
| *** markmcclain has joined #openstack-qa | 18:01 | |
| *** tosky has joined #openstack-qa | 18:02 | |
| *** markmcclain has quit IRC | 18:02 | |
| *** zul has quit IRC | 18:03 | |
| *** markmcclain has joined #openstack-qa | 18:05 | |
| *** dkorn has joined #openstack-qa | 18:06 | |
| *** ildikov has quit IRC | 18:12 | |
| *** openstackgerrit has joined #openstack-qa | 18:13 | |
| *** leseb has joined #openstack-qa | 18:13 | |
| *** leseb has quit IRC | 18:14 | |
| *** leseb has joined #openstack-qa | 18:14 | |
| *** leseb has quit IRC | 18:19 | |
| *** mpavlase has joined #openstack-qa | 18:20 | |
| *** Mike77 has joined #openstack-qa | 18:26 | |
| *** kgriffs is now known as kgriffs|afk | 18:28 | |
| *** mpavlase has quit IRC | 18:30 | |
| *** markmcclain has quit IRC | 18:31 | |
| *** markmcclain has joined #openstack-qa | 18:31 | |
| *** mpavlase has joined #openstack-qa | 18:31 | |
| openstackgerrit | A change was merged to openstack-dev/devstack: remove the build timeouts https://review.openstack.org/96143 | 18:31 |
| openstackgerrit | A change was merged to openstack/tempest: Fix warnings from build_sphinx https://review.openstack.org/97833 | 18:31 |
| *** markmcclain has quit IRC | 18:39 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-dev/devstack: Add flag to set nova v3 api for tempest https://review.openstack.org/98525 | 18:40 |
| *** SridharG has quit IRC | 18:44 | |
| dims_ | mriedem, sdague - have you seen this one? "Instance failed network setup after 1 attempt" AND filename:"logs/new/screen-n-cpu.txt" - 7,177 in 48 hours build_status is failure in all of them | 18:46 |
| dims_ | we have a network_allocate_retries to bump up the number of retries, it's set to 0 currently | 18:47 |
| openstackgerrit | Sabeen proposed a change to openstack/tempest: Basic scenario test - create and delete a stack using a HOT template https://review.openstack.org/97257 | 18:49 |
| mriedem | dims_: log? | 18:50 |
| mriedem | dims_: neutron or nova-network? | 18:50 |
| dims_ | i was looking at the logs in your email | 18:51 |
| dims_ | one sec | 18:51 |
| dims_ | http://logs.openstack.org/39/97139/8/check/check-tempest-dsvm-ironic-postgres-nv/34a2e8c/logs/screen-n-cpu.txt.gz#_2014-06-12_13_13_27_268 | 18:51 |
| mriedem | hmm ironic | 18:52 |
| mriedem | dims_: http://logs.openstack.org/39/97139/8/check/check-tempest-dsvm-ironic-postgres-nv/34a2e8c/logs/screen-ir-cond.txt.gz?level=TRACE#_2014-06-12_12_53_01_559 | 18:52 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Ironic: Add deploy kernel and deploy ramdisk to driver_info https://review.openstack.org/95529 | 18:52 |
| mriedem | MACAlreadyExists: A Port with MAC address fa:16:3e:38:f4:72 already exists. | 18:52 |
| mriedem | dims_: ironic bug | 18:52 |
| mriedem | non-voting job | 18:52 |
| mriedem | so no haven't seen it | 18:52 |
| dims_ | mriedem, i saw it first there, but i see it in other places too http://logs.openstack.org/98/99598/2/check/check-grenade-dsvm/454e4a8/logs/new/screen-n-cpu.txt | 18:53 |
| mriedem | dims_: known issue | 18:54 |
| mriedem | https://bugs.launchpad.net/grenade/+bug/1315095 | 18:54 |
| mriedem | hard to run network things when n-net doesn't start :) | 18:54 |
| mriedem | e-r isn't reporting since it looks like the logstash cluster is down or something, kibana is not working for me | 18:55 |
| dims_ | y me too | 18:55 |
| mriedem | dims_: read scrollback in -infra | 18:57 |
| *** kgriffs|afk is now known as kgriffs | 18:58 | |
| *** devlaps has quit IRC | 18:58 | |
| *** cbader has joined #openstack-qa | 18:58 | |
| *** markmcclain has joined #openstack-qa | 18:59 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add a config option to select compute xml api https://review.openstack.org/99737 | 19:00 |
| *** Mike77 has quit IRC | 19:08 | |
| *** kgriffs is now known as kgriffs|afk | 19:09 | |
| *** kgriffs|afk is now known as kgriffs | 19:10 | |
| mriedem | sdague: this is the new guy i started seeing today https://bugs.launchpad.net/glance/+bug/1329138 | 19:11 |
| sdague | oh, we turned of er for 30 minutes to see if it would impact the indexing | 19:12 |
| mriedem | do we run devstack with ssl compression on? | 19:12 |
| mriedem | wondering about glance api workers | 19:12 |
| mriedem | http://git.openstack.org/cgit/openstack/glance/tree/etc/glance-api.conf#n58 | 19:12 |
| mtreinish | mriedem: I'd say just bump up the worker count anyway it can't hurt | 19:13 |
| mriedem | doing it now | 19:14 |
| mtreinish | dkranz: about your spec that you put on meeting agenda, what about xml? I just assumed the same method would work fine for it. | 19:16 |
| mtreinish | The status codes should be the same for json and xml so you can reuse that part of the schema for the xml clients | 19:16 |
| dkranz | mtreinish: Yes, that was one option. But we do not do schema validation for xml. | 19:17 |
| dkranz | mtreinish: I don't like having half-assed support for xml | 19:17 |
| dkranz | mtreinish: And if we really intended to add such later, way to start would be a little different. | 19:18 |
| dkranz | mtreinish: But I do not believe any one will ever add xml schemas for our apis in tempest. | 19:18 |
| mtreinish | I don't think adding schema validation makes any sense for xml. It's going away | 19:18 |
| dkranz | mtreinish: For all apis? | 19:19 |
| dkranz | mtreinish: I wish it were so but have not seen a proposal except for nova. | 19:19 |
| mtreinish | for nova to start, but everything else will probably eventually follow suit | 19:19 |
| dkranz | mtreinish: To be clear, the other option is to just accept that we lose some response code checking that is currently there for xml | 19:20 |
| mtreinish | yeah, I guess that is always an option | 19:21 |
| dkranz | mtreinish: Dealing with the xml is more work and it doesn't make sense if we really don't care. | 19:21 |
| mtreinish | dkranz: did you see: https://review.openstack.org/99737 | 19:21 |
| dkranz | mtreinish: ANyway, that was what I had wanted to discuss. Add to all xml clients, or do nothing with them. | 19:21 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-dev/devstack: Set glance-api workers equal to number of CPUs https://review.openstack.org/99739 | 19:21 |
| dkranz | mtreinish: No, but I will look now. | 19:21 |
| mtreinish | dkranz: ok, I was just curious, because I just +A'd it | 19:22 |
| *** SridharG has joined #openstack-qa | 19:22 | |
| mtreinish | dkranz: that's just a test for now, but it was relevant to what we're talking about | 19:22 |
| mtreinish | I'll bring it to the ML after I get some timing data | 19:22 |
| dkranz | mtreinish: I see a -2 from you there | 19:22 |
| mtreinish | dkranz: that's actually wip | 19:23 |
| dkranz | mtreinish: So what do you mean you +A'd it? It is your patch. | 19:24 |
| mtreinish | dkranz: I was talking about your spec | 19:24 |
| mtreinish | sorry lag in my response | 19:24 |
| *** markmcclain has quit IRC | 19:25 | |
| *** ildikov has joined #openstack-qa | 19:25 | |
| *** leseb has joined #openstack-qa | 19:25 | |
| dkranz | mtreinish: Oh, thanks. I think you just pasted the wrong number a few lines back. | 19:25 |
| dkranz | mtreinish: I am going to start submitting stuff once we decide on the check xml or ignore it. | 19:25 |
| dkranz | mtreinish: THe nova schema check additions completely ignored xml. | 19:26 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-dev/devstack: Set glance-api workers equal to number of CPUs https://review.openstack.org/99739 | 19:26 |
| dkranz | mtreinish: even for v2 | 19:26 |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add a config option to select compute xml api https://review.openstack.org/99737 | 19:26 |
| mtreinish | dkranz: what was the wrong number? | 19:27 |
| *** dkorn has quit IRC | 19:29 | |
| dkranz | mtreinish: ok, never mind. You were saying two different things about two different patches and I did not realize that. | 19:29 |
| *** leseb has quit IRC | 19:29 | |
| *** cbader has quit IRC | 19:29 | |
| *** pcrews has joined #openstack-qa | 19:31 | |
| pcrews | Hi. Been discussing adding namespaces to elastic-recheck. This is to allow projects to benefit from it without destroying its utility for debugging gate: https://etherpad.openstack.org/p/elastic-recheck-namespace | 19:33 |
| pcrews | Currently there is 'all pipelines', 'gate', and 'uncategorized' - in terms of adding namespaces - are there opinions on having it as a filter on existing dashboards or should a separate tab/dashboard be created like 'filter by namespace'? | 19:34 |
| pcrews | Our goal is to be as unobtrusive as possible for the gate fire-fighters (not breaking anything you guys rely on) | 19:35 |
| sdague | pcrews: my thinking is to consider namespaces as another dimension | 19:36 |
| sdague | so that you'd have maybe something down the left side that would be 'main' (for anything not in another namespace), 'tripleo', '...' | 19:36 |
| sdague | so tripleo would have it's own all queues, gate, and uncategorized | 19:37 |
| openstackgerrit | Emily Hugenbruch proposed a change to openstack/tempest: Changes to add jsonschema for list_servers_detail API. Based required fields on the fields listed in the nova API sample https://review.openstack.org/99742 | 19:37 |
| sdague | because realistically you'll want the uncategorized page | 19:37 |
| pcrews | sdague: ++. That makes sense. | 19:38 |
| sdague | eventually you'll find utility in the others, but uncategorized at least lets you get through all the fails in your backlog to make sure you have fingerprinted them | 19:38 |
| sdague | I guess the only other question is if we add a field to the queries.yaml, or we use directory structure to determine namespaces | 19:39 |
| sdague | it might be better to do it by directories, just because I think we've got 100 bugs currently active | 19:40 |
| pcrews | ah - I had been thinking one could add a field w/ a default to queries.yaml (if field isn't there, it is assumed ''), but you know the code better than I :) | 19:40 |
| dkranz | sdague: Can you respond to the comment from lifeless in https://review.openstack.org/#/c/92497/ ? I think this is an example of what I sent the email about recently. | 19:43 |
| sdague | pcrews: just because there are a *ton* of files, I think it would be mentally easier to do namespaces with directories | 19:44 |
| sdague | realistically they all filter through a load_queries function | 19:44 |
| sdague | so changing that later is actually the easiest part of any of this :) | 19:44 |
| sdague | the knarly part is going to be on the web dashboards, as those are basically cron generated static files that unify on the glass | 19:45 |
| sdague | it's not terrible, but just a little gorpy | 19:45 |
| *** markmcclain has joined #openstack-qa | 19:45 | |
| sdague | pcrews: I was just throwing some implementation hints into the etherpad | 19:46 |
| sdague | dkranz: I think we rolled that all into another set of fixes | 19:47 |
| *** nati_ueno has quit IRC | 19:47 | |
| dkranz | sdague: So should it be abandoned? | 19:47 |
| sdague | dkranz: oh, actually we didn't | 19:47 |
| sdague | after looking at that output I think we decided the start marker wasn't as interesting | 19:47 |
| lifeless | sdague: right, but see my comment. | 19:48 |
| dkranz | sdague: ok, your call what to do. I just saw it and figured it would not show on your dashboard. | 19:48 |
| sdague | dkranz: true | 19:49 |
| sdague | lifeless: I'd say just remove it | 19:49 |
| sdague | because we've got test timeouts, we're not subject to stuck tests, right? | 19:49 |
| lifeless | sdague: they won't break the system | 19:49 |
| lifeless | sdague: but they will be hard to diagnose | 19:49 |
| lifeless | sdague: since the test won't show up in the .testrepository/ log at all at this point | 19:50 |
| lifeless | sdague: because v1 doesn't support concurrency and i haven't finished the migration to v2 for storage | 19:50 |
| sdague | ok, it's something I've not seen before as an issue | 19:50 |
| lifeless | sdague: dkranz is saying he thinks he just saw it. | 19:50 |
| sdague | dkranz: url? | 19:50 |
| sdague | I've got enough contexts juggling I definitely need urls :) | 19:51 |
| dkranz | sdague, lifeless : Sorry, I meant I saw the review with a -1 from sean but his dashboard would not show it as requiring action from him. | 19:51 |
| *** markmcclain has quit IRC | 19:51 | |
| jogo | sdague: https://github.com/openstack/openstack/graphs/commit-activity wow tuesday was productive | 19:51 |
| dkranz | So I was pinging him about it. | 19:52 |
| *** mike77 has joined #openstack-qa | 19:52 | |
| lifeless | dkranz: oh | 19:53 |
| jogo | sdague: so along the lines of dropping our load. what are your thoughts on if a patch fails pep8/unit tests/docs to not run (or just kill) any integration testing? | 19:53 |
| sdague | jogo: killing the tests doesn't help that much, because the nodes were allocated | 19:53 |
| sdague | you'd need to prerun | 19:54 |
| jogo | sdague: we can do that, although I have a hunch that will require some zuul changes | 19:54 |
| jogo | sdague: do you know why so many of our nodes are in deleting? | 19:54 |
| sdague | yes, all the more interesting things we could do will require zuul changes | 19:54 |
| sdague | jogo: go read -infra :) | 19:55 |
| sdague | you have to read at least the last 50 lines of scrollback when you jump in a channel :) | 19:55 |
| *** nati_ueno has joined #openstack-qa | 19:56 | |
| sdague | also, the last gate reset that's going on now is large-ops / havana | 19:56 |
| sdague | jogo: you have the propose to drop that for havana? | 19:56 |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add a config option to select compute xml api https://review.openstack.org/99737 | 19:58 |
| jogo | sdague: not sure how to drop it for just havana | 19:58 |
| jogo | oh i See | 19:59 |
| sdague | yeh, you just set the branch regex | 19:59 |
| sdague | oh... keystone port conflict, our old friend - http://logs.openstack.org/93/96293/2/gate/gate-grenade-dsvm-partial-ncpu/7d5d446/logs/old/screen-key.txt.gz | 20:00 |
| sdague | that's the 2nd reset in the gate right now | 20:00 |
| sdague | so when we hit the ephemeral ports in keystone, do we believe that if we waited and retried, the port would be given up later? | 20:01 |
| sdague | morganfainberg: ^^^ | 20:01 |
| jogo | sdague: to be clear about the delete, it sounds like its a rax issue? | 20:05 |
| jogo | sdague: do we try doing force delete? | 20:05 |
| sdague | jogo: probably better to do that in -infra, that's where the context is | 20:06 |
| *** malini1 has quit IRC | 20:08 | |
| *** leseb has joined #openstack-qa | 20:15 | |
| *** boris-42 has quit IRC | 20:18 | |
| *** morganfainberg has quit IRC | 20:18 | |
| *** boris-42 has joined #openstack-qa | 20:18 | |
| *** morganfainberg has joined #openstack-qa | 20:18 | |
| pcrews | sdague: etherpad ideas are solid to me :) Many thanks for the time + input | 20:18 |
| openstackgerrit | A change was merged to openstack-dev/grenade: ceilometer: sync the DB after the upgrade process https://review.openstack.org/99412 | 20:18 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: remove old queries https://review.openstack.org/99756 | 20:22 |
| sdague | pcrews: no prob, thanks for jumping in | 20:24 |
| *** nati_ueno has quit IRC | 20:28 | |
| *** TravT has joined #openstack-qa | 20:29 | |
| *** openstackgerrit has quit IRC | 20:35 | |
| *** markmcclain has joined #openstack-qa | 20:36 | |
| *** openstackgerrit has joined #openstack-qa | 20:36 | |
| *** __afazekas is now known as afazekas | 20:43 | |
| *** nati_ueno has joined #openstack-qa | 20:44 | |
| *** dims_ has quit IRC | 20:47 | |
| *** ekhugen has quit IRC | 20:53 | |
| *** dkorn has joined #openstack-qa | 21:00 | |
| mriedem | is there a meeting or am i early? | 21:01 |
| mtreinish | mriedem: one more hour | 21:01 |
| mriedem | oh gdi | 21:02 |
| mriedem | well i won't be there then :( | 21:02 |
| mriedem | nova v3 api topic is already in ML | 21:02 |
| mriedem | only other topic i added was multi-node testing for nova migrations | 21:02 |
| mriedem | wondering what needs to be done to make that happen | 21:03 |
| *** dkorn has quit IRC | 21:04 | |
| mtreinish | a gate that isn't resource starved | 21:04 |
| morganfainberg | sdague, re: port conflict | 21:05 |
| mriedem | oh psh | 21:05 |
| morganfainberg | sdague, we a wait-retry might mitigate it even further | 21:05 |
| morganfainberg | sdague, s/we// | 21:05 |
| mtreinish | mriedem: have you ever tried to run the migration tests (we have them but I don't think they've ever been run) on your internal system? | 21:06 |
| mriedem | mtreinish: what migration tests? | 21:06 |
| mriedem | oh | 21:06 |
| mriedem | ha | 21:06 |
| mriedem | no | 21:06 |
| *** nati_ueno has quit IRC | 21:07 | |
| morganfainberg | sdague, i'll put a quick retry into devstack up for review today to help further (port in use for starting keystone). long term, we'll piggyback on port 80/443 but that is a ways out still | 21:07 |
| mriedem | don't have multi-node CI setup | 21:07 |
| mtreinish | mriedem: just point tempest against the deployment you have ci running on... | 21:07 |
| *** gszasz has quit IRC | 21:08 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/tempest: Updated from global requirements https://review.openstack.org/96829 | 21:08 |
| sdague | morganfainberg: the problem is, it's really a 1 in 10k chance | 21:11 |
| sdague | I wanted to know if the theory behind it is the port comes back | 21:11 |
| morganfainberg | sdague, and if a quick retry is 1 in 40k it's even better | 21:11 |
| morganfainberg | right? | 21:11 |
| morganfainberg | sdague, depends on what all is using it | 21:11 |
| sdague | sure, if we believe the thing using it will go away | 21:11 |
| morganfainberg | it could free up in seconds, it could never free up | 21:11 |
| afazekas | mriedem: cold or live migration ? | 21:12 |
| morganfainberg | this is why long term we want to just piggyback on port 80/443 http://<host>/identity/ | 21:12 |
| *** markmcclain has quit IRC | 21:12 | |
| *** markmcclain1 has joined #openstack-qa | 21:12 | |
| morganfainberg | this is a stupid bug and we're caught in the middle because IANA and linux don't agree. | 21:12 |
| morganfainberg | sdague, we could also try and register an exception for that port on running stack.sh, but no guarantees how much help that would be either | 21:13 |
| morganfainberg | sdague, (though I would guess stack.sh putting 35357 as excluded from ephemeral would do 90% more than anything else we can do at this point) | 21:14 |
| mriedem | afazekas: both? | 21:14 |
| sdague | morganfainberg: ok, I'm fine with that as well | 21:14 |
| sdague | you want to propose that? | 21:14 |
| mtreinish | mriedem: btw, the logs from the nightlies all go under here: http://logs.openstack.org/periodic-qa/ | 21:14 |
| morganfainberg | sdague, yeah already looking up what needs to be done | 21:14 |
| afazekas | mriedem: the live migration requires nested guest support or physical machine (or a not ready not hardware accelerated qemu feature) | 21:15 |
| mriedem | mtreinish: cool, sec group rule fails in non-isolated jobs | 21:15 |
| mriedem | http://logs.openstack.org/periodic-qa/periodic-tempest-dsvm-neutron-full-non-isolated-master/7dc3868/console.html | 21:15 |
| mriedem | ugh, can we get swift logs out of syslogd please? | 21:16 |
| mriedem | *syslog | 21:16 |
| mtreinish | mriedem: yeah and on the nova-net ones it looks like we leak floating ips and exhaust the pool | 21:16 |
| openstackgerrit | Fei Long Wang proposed a change to openstack/tempest: Add cases for Swift scenario test https://review.openstack.org/99514 | 21:17 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: remove old queries https://review.openstack.org/99756 | 21:17 |
| mtreinish | mriedem: http://logs.openstack.org/periodic-qa/periodic-tempest-dsvm-full-non-isolated-master/b92b844/console.html#_2014-06-12_08_02_55_875 | 21:17 |
| sdague | mriedem: they don't exist anywhere else | 21:18 |
| sdague | swift doesn't use oslo logging | 21:18 |
| mriedem | i'm just complaining | 21:18 |
| sdague | I'm with you though :) | 21:18 |
| sdague | you can filter out some of it with my latest os-loganalyze | 21:18 |
| mriedem | mtreinish: good call - so we can probably use that to floating ip leak in the ssh nova-net bug we have that we were hitting last week right? | 21:18 |
| sdague | swift-proxy and tgtd are marked as debug | 21:19 |
| mtreinish | mriedem: yeah they might be the same bug or related, but the tenant isolation protects us from the over quota in the gate | 21:20 |
| *** openstackrecheck has quit IRC | 21:20 | |
| mtreinish | heh it also hits the secgroup bug | 21:20 |
| *** openstackrecheck has joined #openstack-qa | 21:20 | |
| openstackgerrit | Steve Baker proposed a change to openstack/tempest: Repeat stack delete 4 times https://review.openstack.org/99766 | 21:20 |
| mriedem | hmm, i thought arosen already had a fix for the security group rule bug | 21:22 |
| mriedem | https://review.openstack.org/#/c/92893/ | 21:23 |
| *** openstackrecheck has quit IRC | 21:23 | |
| *** openstackrecheck has joined #openstack-qa | 21:23 | |
| mriedem | actually "BadRequest: This rule already exists in group 1" is probably just a bug in the test not handling tenant isolation properly | 21:24 |
| mtreinish | yeah that makes sense | 21:25 |
| mriedem | the floating ip over quota is interesting though | 21:25 |
| *** mmaglana has joined #openstack-qa | 21:29 | |
| *** TravT has quit IRC | 21:31 | |
| openstackgerrit | lifeless proposed a change to openstack/tempest: Remove test start output in pretty tracer https://review.openstack.org/92497 | 21:32 |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add total runtime to summary output https://review.openstack.org/99771 | 21:37 |
| mtreinish | sdague: ^^^ something we lost when we moved to using subunit-trace | 21:39 |
| sdague | mtreinish: cool, good catch | 21:39 |
| mtreinish | mriedem: http://logs.openstack.org/periodic-qa/periodic-tempest-dsvm-full-non-isolated-master/b92b844/logs/screen-n-api.txt.gz#_2014-06-12_07_11_12_804 | 21:43 |
| mriedem | mtreinish: already fixed | 21:44 |
| mriedem | mtreinish: https://review.openstack.org/#/c/99212/ | 21:45 |
| mtreinish | ok, I wasn't sure if that was the thing from the other day | 21:45 |
| mriedem | were there like a billion of them? | 21:45 |
| mriedem | then yes | 21:45 |
| mriedem | :) | 21:45 |
| mriedem | speaking of which, we need to get this merged https://review.openstack.org/#/c/98871/ | 21:46 |
| mriedem | heading out for the day, later | 21:47 |
| *** mriedem has quit IRC | 21:48 | |
| *** kgriffs is now known as kgriffs|afk | 21:49 | |
| *** lbragstad has quit IRC | 21:50 | |
| *** masayuki_ has joined #openstack-qa | 21:50 | |
| *** mrodden has quit IRC | 21:51 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: have realtime engine only search recent indexes https://review.openstack.org/99776 | 21:52 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: have realtime engine only search recent indexes https://review.openstack.org/99776 | 21:53 |
| *** dustins has quit IRC | 21:55 | |
| *** rockyg has joined #openstack-qa | 21:55 | |
| *** markmcclain1 has quit IRC | 21:56 | |
| *** markmcclain has joined #openstack-qa | 21:56 | |
| *** dims_ has joined #openstack-qa | 21:57 | |
| *** jamielennox is now known as jamielennox|away | 21:58 | |
| *** Obulpathi has quit IRC | 21:58 | |
| openstackgerrit | Sabeen proposed a change to openstack/tempest: Basic scenario test - create and delete a stack using a HOT template https://review.openstack.org/97257 | 22:00 |
| *** nati_ueno has joined #openstack-qa | 22:03 | |
| *** andreaf has joined #openstack-qa | 22:09 | |
| *** mike77 has quit IRC | 22:10 | |
| *** mrodden has joined #openstack-qa | 22:10 | |
| *** mrodden1 has joined #openstack-qa | 22:11 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-dev/devstack: Reserve port 35357 from the ephemeral range https://review.openstack.org/99779 | 22:12 |
| morganfainberg | sdague, ^ | 22:13 |
| morganfainberg | that should do it, i tried to reserve the port as early as possible in the stack.sh process to ensure we're not already doing something else wonky. | 22:13 |
| morganfainberg | oooh.. actually that wont work *doh* need to be root | 22:13 |
| morganfainberg | *grumble* | 22:14 |
| *** mrodden has quit IRC | 22:15 | |
| sdague | morganfainberg: I think it should also go later | 22:15 |
| morganfainberg | sdague, how far later? | 22:15 |
| sdague | like right before the db config | 22:15 |
| morganfainberg | hhhmm | 22:15 |
| sdague | basically once the actual logic starts happening | 22:16 |
| morganfainberg | i think we might need to be fore apt/yum though | 22:16 |
| sdague | we haven't even setup sudo yet | 22:16 |
| sdague | where you put it | 22:16 |
| morganfainberg | right i just realized that | 22:16 |
| morganfainberg | and am moving it for that reason | 22:16 |
| morganfainberg | ok i'll put it right above the db config | 22:17 |
| morganfainberg | sdague, you're right no where prior to that looks "right" | 22:17 |
| *** andreaf has quit IRC | 22:18 | |
| morganfainberg | i guess i could move this into lib/keystone at this point? | 22:18 |
| *** andreaf has joined #openstack-qa | 22:18 | |
| *** andreaf has quit IRC | 22:19 | |
| sdague | yeh, in lib/keystone probably is better | 22:19 |
| morganfainberg | sdague. ok shuffling this around | 22:19 |
| *** andreaf has joined #openstack-qa | 22:20 | |
| *** giulivo has quit IRC | 22:20 | |
| *** mmaglana has quit IRC | 22:21 | |
| *** mpavlase has quit IRC | 22:21 | |
| openstackgerrit | A change was merged to openstack-dev/devstack: [Fedora] Devstack fails to install in firewalld less system https://review.openstack.org/99047 | 22:23 |
| *** dkranz has quit IRC | 22:24 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-dev/devstack: Reserve port 35357 from the ephemeral range https://review.openstack.org/99779 | 22:24 |
| *** dkranz has joined #openstack-qa | 22:25 | |
| *** markmcclain has quit IRC | 22:27 | |
| *** weshay has quit IRC | 22:28 | |
| *** rbak has quit IRC | 22:35 | |
| *** leseb has quit IRC | 22:36 | |
| *** leseb has joined #openstack-qa | 22:36 | |
| *** andreaf has quit IRC | 22:36 | |
| *** leseb has quit IRC | 22:41 | |
| *** SridharG has quit IRC | 22:44 | |
| *** rbak has joined #openstack-qa | 22:57 | |
| cyeoh | maurosr: so back to the regression testing I think we need (whether periodic or not): | 23:01 |
| cyeoh | v2 (obviously) | 23:01 |
| cyeoh | v2.1 (when we have it) | 23:01 |
| *** dims_ has quit IRC | 23:01 | |
| cyeoh | v2.1microversions (as we roll out new api features) | 23:01 |
| *** aimon has quit IRC | 23:02 | |
| cyeoh | v3 features (which gets smaller as v2.1microversions gets bigger) | 23:02 |
| sdague | cyeoh: so I actually think that the combinatorics of microversions are going to mean that tempest can't cover the whole space | 23:02 |
| sdague | and we need to figure out how to be more rigourous on the nova tree side | 23:03 |
| cyeoh | sdague: yep, I agree, especially with scenario testing | 23:03 |
| cyeoh | going to make some hard decisions about various points in microversions where we test really well | 23:03 |
| *** rockyg has quit IRC | 23:03 | |
| sdague | well you can test it all on the nova side | 23:04 |
| maurosr | so the fact that we're proposing tests for '10% of unstability' api, doesn't mean that we're experimenting imho | 23:04 |
| dkranz | cyeoh: We need to get to the point where complete functional nova testing happens only on nova commits | 23:04 |
| cyeoh | not in an integrated way we can't test on the nova side though can we? | 23:04 |
| maurosr | I would call it evolving, as sometime in the past we didn't cover much of the nova's api and now we do | 23:04 |
| *** adalbas has quit IRC | 23:04 | |
| dkranz | cyeoh: No but I think we will have to give up on that. | 23:05 |
| *** jamielennox|away is now known as jamielennox | 23:05 | |
| dkranz | cyeoh: We can't run a complete functional test of every project on every commit to every other project | 23:05 |
| sdague | dkranz: right, of every surface | 23:05 |
| cyeoh | dkranz: well we can certainly get better at unittests on the nova side - and from the api I've certainly been pushing for much better api sample tests | 23:05 |
| dkranz | cyeoh: We have gotten away with this thus far due to incomplete tests and small number of projects | 23:06 |
| dkranz | cyeoh: But it does not scale | 23:06 |
| sdague | cyeoh: so I think we should actually do a nova functional test job | 23:06 |
| cyeoh | which starts to overlap a bit on what some of the tempest stuff does | 23:06 |
| sdague | that starts real nova, on a fake driver, on devstack | 23:06 |
| sdague | and tests that way | 23:06 |
| sdague | that only runs on nova | 23:06 |
| sdague | on nova commits | 23:06 |
| sdague | which is actually the path that neutron is headed | 23:06 |
| cyeoh | sdague: ah interesting. I'd like to see that, don't know I can volunteer the bandwidth for it :-) | 23:06 |
| sdague | and swift already has | 23:07 |
| maurosr | cyeoh: I can ;) | 23:07 |
| cyeoh | its goes a bit further than what the api sample tests do | 23:07 |
| cyeoh | maurosr: cool :-) | 23:07 |
| sdague | yeh, realistically, I think the api samples tests have possibly outlived their usefulness | 23:07 |
| cyeoh | sdague: well they do pick up a lot of bugs that the unittests don't but they're definitely replaceable | 23:08 |
| sdague | right, I think this would be a better replacement | 23:08 |
| cyeoh | and once we get v2.1 and jsonschema etc all in they won't be needed for docs anymore | 23:08 |
| sdague | because it would be kind of uber mock | 23:08 |
| sdague | where just the driver is faked out | 23:08 |
| sdague | which would also mean you could go for fault injection testing as well | 23:09 |
| sdague | it should go pretty fast | 23:09 |
| *** dkranz is now known as dkranz_afk | 23:09 | |
| sdague | it's "sort of" what largeops is doing | 23:09 |
| cyeoh | so I guess that brings up the question if tempest eventually will even need to do stable API testing and just do scenario testing instead? | 23:10 |
| cyeoh | (except perhaps for cert related reasons?) | 23:10 |
| maurosr | cyeoh: question, by your experience on nova-v3 compared to v2, is there many things that need to be testing beyond parameters rename, resource endpoint rename, and status code changes? | 23:10 |
| cyeoh | maurosr: there's a big list of v2->v3 changes somewhere. Will have to dig it up though | 23:10 |
| cyeoh | maurosr: but remember its a totally different codebase | 23:11 |
| sdague | cyeoh: I think we'll need a representative part of the API | 23:11 |
| stevebaker | oh, I wanted to talk fault-injection. I want to write heat tests which test resilience to transient failures in the rest of the cloud. Are there options now for injecting those failures? | 23:11 |
| sdague | stevebaker: nope | 23:12 |
| stevebaker | (; | 23:12 |
| stevebaker | ); | 23:12 |
| maurosr | yeah, but tempest is kind of a user.. so if doest go much different from that, json schema are awesome source to test even with micro versioning (if we have a new schema for new micrversions) | 23:12 |
| * afazekas Would be nice if it would be approved before I go to sleep https://review.openstack.org/#/c/99694/ (01:11 AM) | 23:12 | |
| sdague | stevebaker: we get plenty of faults in the normal heat tests. :) we don't need to inject more yet... | 23:12 |
| stevebaker | afazekas: thanks btw | 23:12 |
| sdague | in seriousness though, I have some ideas for this in K | 23:12 |
| afazekas | stevebaker: can you ping mkollaro, she may know how to add it to destroystack | 23:13 |
| cyeoh | sdague: what do you mean by representative part and what is the advantage of just testing a subset? | 23:13 |
| sdague | but there is no way of getting near it until then | 23:13 |
| stevebaker | sdague: lol, I know. It would be nice to be able to inject a specific failure for a specific operation though | 23:13 |
| sdague | stevebaker: yep, I'm with you | 23:13 |
| cyeoh | for cert purposes we'd need to test all. If we have good testing inside nova then we don't need just isolated API tests at all right? | 23:13 |
| sdague | I just don't think it's a priority yet, as we've got tons of bugs unresolved under the current methods | 23:14 |
| cyeoh | as scenario testing would effectively cover a lot of the API anyway | 23:14 |
| sdague | cyeoh: well, scenario testing goes through the official clients, so it doesn't really | 23:14 |
| sdague | because the clients mask api compat issues | 23:14 |
| sdague | honestly, I don't know where exactly we'll end up | 23:15 |
| cyeoh | ok | 23:16 |
| sdague | we keep kind of steering this ship in the most useful direction for the next six months. And right now I think unless we got a ton of new tempest devs and reviewers, having it cover all the micro version scope isn't going to be successful | 23:16 |
| cyeoh | oh yea on that I agree. Its the big downside to going the microversion route, rather than big bang version increments | 23:16 |
| sdague | because with more folks using this code, usability is probably a higher priority right now than coverage growth | 23:17 |
| sdague | especially where usability == debugability | 23:17 |
| cyeoh | sure, agreed. | 23:17 |
| sdague | as this gets rolled out in defcore, that's going to drive a lot of issues about how this maps in real clouds | 23:18 |
| *** markmcclain has joined #openstack-qa | 23:18 | |
| mkollaro | afazekas: what's the talk about? | 23:18 |
| stevebaker | mkollaro: hi | 23:19 |
| afazekas | fault-injection | 23:19 |
| mkollaro | stevebaker: hello | 23:19 |
| mkollaro | stevebaker: what kind of fault injections would you like to do? | 23:20 |
| stevebaker | mkollaro: I'm working on making heat resilient to transient cloud failures. destroystack might be a good place to write tests which check that | 23:20 |
| mkollaro | stevebaker: yep, that's what it's designed for...could you give me some example of a test scenario? | 23:21 |
| stevebaker | mkollaro: APIs not responding, APIs returning 500s, nova servers coming up in and ERROR state. That would cover most scenarios | 23:21 |
| mkollaro | stevebaker: I hope heat is a bit smarter about error handling, because I'm trying to write nova scheduler tests and I just found out nova is stupid.... | 23:21 |
| stevebaker | mkollaro: ahem, heat is terrible at this ;) | 23:21 |
| mkollaro | everything is terrible at this... | 23:22 |
| stevebaker | mkollaro: so is nova. I killed n-api and ended up with an undeletable server :/ | 23:22 |
| mkollaro | yeah, I thought it would at least survive a compute restart...but no, in havana it just gets stuck | 23:22 |
| mkollaro | very difficult to write fault injection tests when nothing is fault tolerant | 23:23 |
| stevebaker | mkollaro: test scenario. heat creates a server, server goes to ERROR, heat does delete and create retries until a server stands up (some of the deletes might fail too) | 23:23 |
| mkollaro | hm, I wonder how we could simulate the error state of the VM...well, we could just restart the compute node where it was spawning | 23:24 |
| stevebaker | I was thinking swapping out a glance image with a bogus one | 23:24 |
| mkollaro | maybe we could kinda jump into nova internals and force it to fail | 23:25 |
| mkollaro | or create a mock nova api that would simply say it failed, but that wouldn't be so realistic | 23:25 |
| stevebaker | no, it would be nice if it was a real cloud | 23:26 |
| mkollaro | yep | 23:26 |
| mkollaro | I don't know how to trigger the deletion errors though | 23:26 |
| mkollaro | but some nova devels might help us figuring that one out | 23:27 |
| *** atiwari has quit IRC | 23:27 | |
| mkollaro | so, heat is supposed to do re-tries when the created VM is not accessible? | 23:27 |
| stevebaker | mkollaro: can you block access to an API in a way other than killing it? | 23:28 |
| mkollaro | e.g., when the network isn't working properly and such? | 23:28 |
| afazekas | nova boots anything as raw image, may be it can fail with a wrong qcow2 | 23:28 |
| sdague | stevebaker: why don't you do that part via mocks? | 23:28 |
| sdague | like feedback the client api calls that would be bad | 23:28 |
| sdague | because your view of the world is through the clients | 23:28 |
| stevebaker | sdague: it is | 23:29 |
| *** danfinch1 is now known as danfinch | 23:29 | |
| sdague | that would get you something today that would be down that path | 23:29 |
| sdague | not quite the same as live, but you could probably control a lot better | 23:29 |
| stevebaker | sdague: I'm working on client plugins, we could have test plugins which inject errors | 23:30 |
| mkollaro | creating the actual fault that would realistically happen might be more simple than creating the mock APIs | 23:30 |
| mkollaro | plus it might reveal some extra errors | 23:30 |
| sdague | mkollaro: it depends. Right now, not in the gate. | 23:31 |
| mkollaro | sdague: yeah, I doubt these would be ever used in the gate, it's quite slow and requires resources | 23:31 |
| *** mmaglana has joined #openstack-qa | 23:31 | |
| sdague | I actually think we will do fault injection there down the road | 23:31 |
| mkollaro | just the swift replication tests take 2 minutes each (when they succeed), plus I reset the state of openstack by creating snapshots and restoring them, which again takes time | 23:32 |
| sdague | but right now, with 18hr merge queue, we really need more eyes on the existing bugs instead of injecting new ones :) | 23:32 |
| stevebaker | pff | 23:32 |
| sdague | anyway, dinner time. Night all. | 23:34 |
| mkollaro | good night | 23:34 |
| afazekas | good night | 23:35 |
| *** mmaglana has quit IRC | 23:36 | |
| mkollaro | stevebaker: ping me tomorrow, we can come up with some test plan :) | 23:39 |
| *** tosky has quit IRC | 23:40 | |
| stevebaker | mkollaro: yeah, I'll need to think about this some more | 23:40 |
| *** GMann has joined #openstack-qa | 23:40 | |
| *** GMann is now known as GMann_TGIP | 23:43 | |
| *** GMann_TGIP is now known as GMAnn_TGIF | 23:43 | |
| *** GMAnn_TGIF is now known as GMann_TGIF | 23:44 | |
| *** leseb has joined #openstack-qa | 23:47 | |
| *** leseb has quit IRC | 23:51 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!